AI API Cost Calculator

Compare pricing across OpenAI, Claude, Gemini, DeepSeek, and Mistral. Calculate your monthly AI spending in seconds.

📊 Calculate Your Monthly Cost

Total API calls per day

~4 chars per token (200 tokens ≈ 1 paragraph)

Generated response length

📋 Full API Pricing Reference

Provider Model Input $/1M Output $/1M Context Tier
🤖 OpenAI GPT-5.4 $2.50 $15.00 270K flagship
🤖 OpenAI GPT-5.4 mini $0.75 $4.50 270K mid
🤖 OpenAI GPT-5.4 nano $0.20 $1.25 270K budget
🤖 OpenAI GPT-5.2 $1.75 $14.00 200K flagship
🤖 OpenAI GPT-5.2 Pro $21.00 $168.00 200K premium
🤖 OpenAI o3 (reasoning) $10.00 $40.00 200K premium
🤖 OpenAI o4-mini (reasoning) $1.10 $4.40 200K mid
🧠 Anthropic Claude Opus 4.6 $5.00 $25.00 1M premium
🧠 Anthropic Claude Sonnet 4.6 $3.00 $15.00 200K flagship
🧠 Anthropic Claude 3.5 Haiku $0.80 $4.00 200K mid
💎 Google Gemini 2.5 Pro $2.00 $12.00 1M flagship
💎 Google Gemini 2.5 Flash $0.30 $2.50 1M mid
💎 Google Gemini 2.0 Flash-Lite $0.07 $0.30 1M budget
🔥 DeepSeek DeepSeek V3.2 $0.28 $0.42 128K budget
🔥 DeepSeek DeepSeek R1 (reasoning) $0.55 $2.19 128K mid
🌬️ Mistral Mistral Large $2.00 $6.00 128K flagship
🌬️ Mistral Mistral Small $0.20 $0.60 128K budget
🌬️ Mistral Mistral Nemo $0.01 $0.01 128K budget
🦙 Meta (hosted) Llama 4 Maverick $0.15 $0.60 1M budget
🦙 Meta (hosted) Llama 4 Scout $0.08 $0.30 10M budget

Sources: openai.com, anthropic.com, ai.google.dev, deepseek.com, mistral.ai — April 2026

❓ Frequently Asked Questions

What is the cheapest LLM API in 2026?
As of April 2026, the cheapest capable LLM APIs are Mistral Nemo ($0.015/1M tokens) and Llama 4 Scout ($0.08/1M input). For models with strong general capability, DeepSeek V3.2 ($0.28 input, $0.42 output) offers the best price-to-performance ratio.
How much does it cost to run a GPT-4 class chatbot?
A chatbot handling 1,000 users/day with 20 messages each costs approximately $90-450/month on GPT-5.4 mini, or $2,700-13,500/month on GPT-5.4. Using DeepSeek V3.2 would cost just $5-25/month for the same workload.
OpenAI vs Claude vs Gemini: which is cheapest?
At the budget tier, Google Gemini 2.0 Flash-Lite ($0.075/$0.30) is cheapest. At the flagship tier, DeepSeek V3.2 ($0.28/$0.42) dominates. For premium quality, Claude Opus 4.6 ($5/$25) is cheaper than GPT-5.4 ($2.50/$15) on output but more expensive on input.
How do I reduce my AI API costs?
Key strategies: (1) Use cheaper models for simple tasks (routing/classification), (2) Enable prompt caching to reduce input costs by up to 90%, (3) Use Batch API for non-real-time workloads (50% discount at OpenAI), (4) Optimize prompt length, (5) Consider self-hosted open-source models for high-volume workloads.